A numerical study of strong-post double-faced W-beam and Thrie-beam guardrails under impacts of vehicles of multiple size classes

https://doi.org/10.1016/j.aap.2021.106286Get rights and content

Highlights

  • Developed FE models of double-faced W-beam and Thrie-beam guardrails.

  • Performed nonlinear FE simulations of guardrails impacted by four different types of vehicles.

  • Evaluated performance of double-faced W-beam and Thrie-beam guardrails at differentheights.

Abstract

Median barrier systems are safety features between opposite travel lanes to redirect or prevent errant vehicles from intruding into oncoming traffic lanes. Different barriers systems have been developed and used for decades. In this study, finite element (FE) modeling and simulations were adopted to study the performance of double-faced W-beam and Thrie-beam guardrails at 29- and 31-inch installation heights. The in-service guardrail performance was evaluated under impacts of multiple sized vehicles at Test Level 4 and Test Level 5 conditions specified in Manual for Assessing Safety Hardware (MASH). The effectiveness of the guardrails was assessed using guardrail dynamic deflections, vehicle responses, and vehicle redirection characteristics. The MASH exit box criterion, MASH evaluation criterion F (roll and pitch angle limit), and MASH evaluation criterion N (vehicle trajectory) were adopted in the evaluations. Additionally, occupant safety and injury risk were determined using occupant impact velocities (OIVs), occupant ridedown accelerations (ORAs), and acceleration severity indices (ASIs). The crash simulation results showed that both W-beam and Thrie-beam guardrails could retain all test vehicles and prevent them from getting into oncoming travel lanes. All the guardrails were considered generally effective in terms of occupant risk factors and vehicle impact responses. It was also observed that in certain cases, the installation height and the type of guardrail blockout could affect the impact severity for small-sized vehicles.

Introduction

Median barrier systems are essential highway safety features serving the purpose of safely containing and redirecting errant vehicles to prevent them from entering oncoming travel lanes or dangerous areas. Various types of barriers and guardrails are available, and they can be categorized into three major groups: rigid (e.g., concrete barriers), semi-rigid (e.g., W-beam and Thrie-beam guardrails), and flexible (e.g., cable barriers) systems. Semi-rigid barriers, including different types of W-beam and Thrie-beam guardrails, are the most commonly used safety devices on U.S. highways (Markets and Markets, 2015). When used as median barriers, W-beam and Thrie-beam guardrails are usually installed as two separate lines of single-faced guardrails, with one on each side of the median. Such installations often cause difficulty for maintenance such as repair, mowing and vegetation management. A practical solution is to use a single line of double-faced guardrail by connecting two rails on the same posts with an offset by blockouts (see Fig. 1) (Fang et al., 2013).

In 1993, Ross et al. developed the first uniform procedures for assessing the safety performance of roadside hardware including median barriers, which were adopted as the standard guideline, known as the NCHRP Report 350 (Ross et al., 1993), until it was replaced by the new standard, Manual for Assessing Safety Hardware (MASH) (Transportation Officials, 2009). Most of the studies in literature on the analysis and design of roadside safety barriers, including both full-scale crash tests and numerical simulations, were conducted at the popular test level, i.e., Test Level 3 (TL-3) that using 1100C and 2270P test vehicles at the speed of 44 mph with a 25° impact angle. Research on barrier performance at higher test levels, i.e., Test Level 4 (TL-4) and Test Level 5 (TL-5), are quite limited, especially those considering multiple crash scenarios using different sized vehicles (Meng and Untaroiu, 2020, Meng et al., 2020). MASH TL-4 has the same test conditions as TL-3, except that an additional 10,000S test vehicle is used with an impact speed of 56 mph (90 km/h) and a 15° impact angle. MASH TL-5 has the same test conditions as TL-3 but with an additional 36,000 V test vehicle at an impact speed of 50 mph (80 km/h) and a 15° impact angle. This study was intended to fill the gap of evaluating the in-service performance of strong-post double-faced guardrails under impacts of multiple sized vehicles at higher test levels.

The earliest testing involving Thrie-beam guardrails was conducted by Bryden and Hahn in 1981 to validate the usability of a ten-gauge Thrie-beam guardrail for bridge rails (Bryden and Hahn, 1981). They conducted a series of crash tests of vehicles impacting corrugated steel Thrie-beam guardrails to determine rail deflections, structural adequacy, vehicle deaccelerations, and vehicle damages. In 2001, Ray and Weir conducted performance evaluation of four in-service guardrail systems: the G1 cable guiderail, G2 weak-post W-beam guardrail, and the G4(1S) and G4(1 W) strong-post W-beam guardrails (Ray and Weir, 2001). In 2005, Gabler et al. evaluated a modified Thrie-beam guardrail system; they concluded that the Thrie-beam guardrail was capable of containing and redirecting passenger vehicles as well as a limited number of heavy vehicles (Gabler et al., 2005). In a subsequent study by Gabler et al. (Gabler et al., 2006), they investigated fatalities and injuries in accidents involving W-beam guardrails in New Jersey and found that the guardrails generally performed well in vehicular crashes. There are over half of the fatal collisions involved secondary events, i.e., either a second impact or a rollover. Hampton et al. conducted crash tests and finite element (FE) analysis of the G4(1S) W-beam guardrail with existing damaged sections (Hampton et al., 2010). They concluded that a prescribed deflection of 0.279 m or more on the post and rail would result in vehicle vaulting over the guardrail. In the work by Alluri et al., they evaluated the safety performance of the G4(1S) guardrail system whose effectiveness was measured by the percentage of vehicles prevented from crossing the guardrail during crashes (Alluri et al., 2012). They found that the guardrail had a crossover prevention rate of 78% for medium and heavy trucks, which was significantly lower than the prevention rates of 97.5% for cars and 91.6% for light trucks. In 2013, researchers at the Midwest Roadside Safety Facility studied the safety performance of the Midwest Guardrail System with no blockouts; the new system was successfully tested under MASH TL-3 conditions (Schrum et al., 2013).

With the rapid development in computational mechanics in the past two decades, FE modeling and simulations of vehicular crashes have been increasingly used in roadside safety research. Most of the current FE models of vehicles and roadside safety structures in the public domain were initially developed at the National Crash Analysis Center (NCAC). Kan et al. developed an integrated FE model that included a vehicular structure, detailed interior components, a Hybrid III dummy, seatbelt, and airbag for crashworthiness evaluation (Kan et al., 2001). Mohan et al. improved and validated a previously developed model of a 1996 Ford F800 single-unit truck that was used as the standard TL-4 vehicle in NCHRP Report 350 (Mohan et al., 2007). Marzougui et al. developed the FE model of a W-beam guardrail and validated it using full-scale crash test data (Marzougui et al., 2007). The guardrail model was shown to provide an accurate representation of the real barrier system based on roll and yaw angles of the vehicle. Plaxico et al. investigated the failure mechanism of the bolted connection of a W-beam rail to a guardrail post, which could have a significant effect on the performance of a guardrail system (Plaxico et al., 2003). Atahan studied the performance of a strong-post W-beam guardrail using a full-scale crash test at NCHRP Report 350 Test 3-11 conditions (Atahan, 2002). Upon identifying the cause of failure, they developed a new W-beam guardrail with improved performance using FE simulations. Marzougui et al. modified six G9 Thrie-beam guardrails and three G4(1S) W-beam guardrails using FE simulations to satisfy MASH TL-3 conditions (Marzougui et al., 2012). Hampton et al. performed FE simulations to evaluate the performance of strong-post W-beam guardrails with missing posts under impact conditions specified by NCHRP Report 350 (Hampton and Gabler, 2013). The simulation results showed that guardrails with even one missing post could have a remarkably decreased performance under vehicular impacts due to tire snagging. Numerical modeling and simulations were shown to be a viable means and employed intensively in other roadside safety studies (Ray, 1996, Atahan, 2010, Ray and Patzner, 1997, Patzner et al., 1999, Ray et al., 2004, Reid, 2004, Whitworth et al., 2004, Bligh et al., 2004, Ferdous et al., 2011).

The objective of this study was to identify weakness of these guardrails and provide insight into how to potentially improve their designs and performance. In addition to the commonly used passenger vehicles for TL-3 conditions, a single-unit truck and a tractor-trailer were considered in this study for TL-4 and TL-5 conditions, respectively. The modeling and simulation work was conducted on two double-faced W-beam guardrails (with placement heights of 29 and 31 in.) and four Thrie-beam guardrails (with wood blockouts and steel blockouts, and each with placement heights of 29 and 31 in.).

Section snippets

FE modeling of test vehicles

Full-scale crash tests at MASH TL-4 and TL-5 conditions require a total of four types of vehicles: a small passenger car (1100C), a pickup truck (2270P), a single-unit truck (10,000S), and a tractor trailer (36,000 V). In the vehicular crash simulations of this study, a 1996 Dodge Neon, a 2006 Ford F250, a 1996 Ford F800, and a 1991 GMC day-cab tractor-trailer were adopted (see Fig. 2), corresponding to the MASH 1100C, 2270P, 10,000S, and 36,000 V vehicles, respectively. The specifications of

MASH TL-4 and TL-5 conditions

The W-beam guardrail and two Thrie-beam guardrails were evaluated under MASH TL-4 and TL-5 conditions at 29- and 31-inch installation heights. At TL-4 conditions, as given in Table 4, the guardrails were impacted by 1100C and 2270P vehicles at 62 mph (100 km/h) with a 25° impact angle. Additionally, the guardrails were also impacted by the 10,000S vehicle at 56 mph (90 km/h) with a 15° impact angle. At TL-5 conditions, as given in Table 5, the guardrails were impacted by the 1100C and 2270P

Vehicular responses

A total of 24 simulations were conducted using FE analyses for the two double-faced W-beam guardrails and the four Thrie-beam guardrails, as given by the simulation matrix in Table 6. The overall vehicular responses in terms of vehicle redirection characteristics and guardrail engagement were defined by one of following types of responses for all the simulation results.

  • 1)

    Redirected (R): The vehicle was safely redirected and the MASH exit box criterion was met.

  • 2)

    Redirected but failed to pass the

Concluding remarks

In this study, the in-service performance of double-faced W-beam and Thrie-beam guardrails was evaluated using full-scale FE simulations under impacts of multiple sized vehicles. The commonly used installation heights, i.e., 29 and 31 in., were considered for all guardrails and two blockout types, i.e., wood and steel blockouts were adopted for Thrie-beam guardrails. In addition to vehicle trajectories and guardrail engagement, the guardrail performance was also evaluated using OIV, ORA, and

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

The authors greatly appreciate the support for this study by the North Carolina Department of Transportation (NCDOT) under Project No. 2015-10.

References (48)

  • A.O. Atahan

    Vehicle crash test simulation of roadside hardware using LS-DYNA: a literature review

    Int. J. Heavy Veh. Syst.

    (2010)
  • Bligh, R.P., Abu-Odeh, A.Y., Hamilton, M. E., Seckinger, N.R., 2004. Evaluation of Roadside Safety Devices Using Finite...
  • J.E. Bryden et al.

    Crash tests of light-post Thrie-beam traffic barriers

    Transp. Res. Rec.

    (1981)
  • Buth, C.E., Menges, W.L., Butler, B.G., 1999. NCHRP Report 350 Test 3/21 of the Vertical Flared Back Transition (No....
  • Committee for European Norms, 2010. Road Restraint Systems Part 2: Performance Classes, Impact Test Acceptance Criteria...
  • Fang, H., Li, N., Tian, N., 2010. Median Barrier Placement on Six-lane, 46-foot Median Divided Freeways. NCDOT Project...
  • Fang, H., Gutowski, M., Li, N., DiSogra, M., 2013. Performance Evaluation of NCDOT W-beam Guardrails under MASH TL-2...
  • H. Fang et al.

    Numerical simulation of single-and double-faced W-beam guardrails under vehicular impacts on six-lane divided freeways

    WIT Trans. Eng. Sci.

    (2013)
  • H. Fang et al.

    Performance Evaluation of 29-inch and 31-inch W-beam Guardrails on Six-lane, 46-foot Median Divided Freeways

  • M.R. Ferdous et al.

    Performance limit analysis for common roadside and median barriers using LS-DYNA

    Int. J. Crashworthiness

    (2011)
  • Gabler, H.C., Gabauer, D. J., Bowen, D., Chmiel, A., 2005. Evaluation of Cross Median Crashes (No. FHWA-NJ-2005-04)....
  • Gabler, H. C., Gabauer, D. J., Szalaj, W.L., 2006. Safety Audit of Fatalities and Injuries Involving Guide Rail (No....
  • C.E. Hampton et al.

    Limits of acceptable rail-and-post deflection in crash-damaged strong-post W-beam guardrail

    Transp. Res. Rec.

    (2010)
  • C.E. Hampton et al.

    Development of a missing post repair guideline for longitudinal barrier crash safety

    J. Transp. Eng.

    (2013)
  • Cited by (4)

    View full text